CN106354805A - Optimization method and system for searching and caching distribution storage system NoSQL - Google Patents
Optimization method and system for searching and caching distribution storage system NoSQL Download PDFInfo
- Publication number
- CN106354805A CN106354805A CN201610744493.9A CN201610744493A CN106354805A CN 106354805 A CN106354805 A CN 106354805A CN 201610744493 A CN201610744493 A CN 201610744493A CN 106354805 A CN106354805 A CN 106354805A
- Authority
- CN
- China
- Prior art keywords
- ssd
- local ssd
- described local
- target data
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2471—Distributed queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Abstract
The invention provides an optimization method and a system for searching and caching a distribution storage system NoSQL. The method comprises the steps of: S101, presetting local SSD as read-only cache in HBase client; S102, judging whether the target data is located on local SSD or not when the target data is read in HBase client; if so, entering S104, if not, entering S103; reading the target data from HDFS colony and caching to the local SSD, S104, returning the target data by local SSD. The above technical scheme is introduced to local SSD and read-only cache function can be provided; the random reading performance of SSD is developed completely; related document is cached in SSD; the accessing data has locality, the reading HDFS I/O number can be reduced effectively by SSD caching document, and then the performance of distribution storage system is improved, and the purpose of improving data searching efficiency is achieved.
Description
Technical field
The invention belongs to field of computer technology, especially relate to a kind of distributed memory system nosql search caching
Optimization method and system.
Background technology
Single distributed memory system cannot meet the need of storage, management and the search of extensive mass data now
Ask.At present distributed memory system and multiple data centers storage is the technological approaches of the key meeting eb DBMS storage demand.
How to rapidly search for meeting from large-scale data set user's requirement data be current across data center storage system urgently
The appearance of the new storage medium such as problem to be solved, particularly ssd is searched to distributed memory system with its excellent performance
Strap carrys out far-reaching influence, therefore in order to improve search efficiency, needs to break through the search skill based on ssd and load locality characteristic
The bottleneck problem of art.
Relatively conventional mechanical hard disk, solid state hard disc ssd has very big advantage in performance.Ssd has high performance,
It is especially suitable for requiring the request high compared with fast-response time and read-write number of times per second.Ssd applies in large-scale storage systems at present
It is divided into ssd Bedding storage technology and ssd caching technology.Ssd Bedding storage technology can store system by the method for Bedding storage
The high-throughput of system and low access response time, but the difficulty using the method for demixing technology is how to judge different storages
The value of data and file, thus obtaining good throughput and low time delay, needs more ssd with layered tier method,
Need higher carrying cost;Ssd caching technology is temporal locality according to data access and spatial locality adopts ssd
Caching technology as storage system.
The nosql storage system with bigtable as representative common at present, including hbase, cassandra etc., bottom
Group organization data by the way of lsm tree, data write is no longer changed.But the universal solution such as the buffer memory for ssd,
Do not set up specific ssd caching method for the specific load characteristic in upper strata, be not directed to above-mentioned Write once and read multiple yet
Nosql system set up the caching method of data search, especially under strange land environmental condition in data center environment for
The data access of distributed memory system nosql.Corresponding, lead to its data search less efficient, have larger offer to improve
Space.
Content of the invention
In view of this, the embodiment of the present invention provide a kind of distributed memory system nosql search caching optimization method and
System, is not directed to the specific load characteristic in upper strata and distributed memory system nosql Write once and read to solve prior art
The caching method of multiple feature leads to the less efficient technical problem of data search, reaches the mesh improving data search efficiency
's.
The technical scheme that the present invention provides is as follows:
A kind of optimization method of distributed memory system nosql search caching, comprising:
Step s101, in hbase client, pre-sets local ssd as only read buffer;
Step s102, when hbase client reads target data, judges whether target data is located at described local ssd
On, if it is, entering step s104;If it is not, then entering step s103;
Step s103, reads target data from hdfs cluster and is cached on described local ssd;
Step s104, described local ssd returns described target data.
Preferably, before described step s102, also including:
Hbase client sends read requests to corresponding hregionserve;
Hregionserver, according to the read requests receiving, judges whether target data is located locally in internal memory, if
It is then to return described target data from local memory, if it is not, then entering step s102.
Preferably, described step s104 includes:
Described local ssd returns described target data to described local memory, and described local memory returns described mesh
Mark data.
Preferably, before described step s103, also including:
Judge whether the residual memory space of described local ssd meets preset requirement, if it is, entering step s103;
If it is not, then algorithm is replaced in execution, replace the data with existing in described local ssd with the target data reading.
Preferably, methods described also includes:
Hbase client generates compact operation requests, judges compact operation requests corresponding target hfile file
Whether it is located on described local ssd, if it is, execution compact operation;If it is not, then reading target from hdfs cluster
Hfile file cache and executes compact operation, after the completion of compact operation, by former hfile file on described local ssd
Delete from local ssd, and by newly-generated hfile file write hdfs cluster and be cached in local ssd.
Preferably, methods described also includes:
Hbase client generates split operation requests, whether judges split operation requests corresponding target hfile file
On described local ssd, if it is, execution split operation;If it is not, then reading target hfile literary composition from hdfs cluster
Part is cached on described local ssd and executes split operation, after the completion of split operation, by former hfile file from local ssd
Delete.
Corresponding to said method, present invention also offers a kind of optimization system of distributed memory system nosql search caching
System, comprising:
Local ssd, for as read-only buffer setting in hbase client;
First judge module, described for when hbase client reads target data, judging whether target data is located at
On local ssd;
Data read module, for when target data is not located on described local ssd, reading target from hdfs cluster
Data is simultaneously cached on described local ssd;
Data returns module, for when target data is located on described local ssd, returning described from described local ssd
Target data.
Preferably, described system, also include:
Second judge module, whether the residual memory space for judging described local ssd meets preset requirement;
Replacement module, when being unsatisfactory for preset requirement for the residual memory space in described local ssd, execution is replaced and is calculated
Method, replaces the data with existing in described local ssd with the target data reading.
Preferably, described system, also include:
Compact module, for when target hfile file is located on described local ssd, execution compact operates, will
Former hfile file is deleted from local ssd, and by newly-generated hfile file write hdfs cluster and is cached to local ssd
In.
Preferably, described system, also include:
Split module, for when target hfile file is located on described local ssd, execution split operates, will be former
Hfile file is deleted from local ssd.
Using technique scheme, the present invention at least can obtain following technique effects:
The technical scheme that the present invention provides, introducing local ssd provides read-only caching function, gives full play to the random write of ssd
Take performance, ssd caches associated documents, the data because accessing has locality, can be effective by the file caching in ssd
Minimizing read hdfs i/o number, thus improving the performance of distributed memory system, reach improve data search efficiency etc.
Purpose.
In the application, for hbase, by the way of lsm tree, group organization data and upper strata reading data are many for technique scheme
Load, using local ssd as the read-only caching function of hbase cluster, make full use of the feature of hbase itself, especially suitable
In the application such as upper strata remote sensing satellite, data access is read to write few load characteristic more.
Brief description
The optimization method flow chart of the distributed memory system nosql search caching that Fig. 1 provides for embodiment one;
Hbase after distributed memory system after the local ssd of addition that Fig. 2 provides for embodiment one
Hrregionserver configuration diagram;
The compact flow chart that Fig. 3 provides for embodiment two;
The split flow chart that Fig. 4 provides for embodiment three;
The optimization system composition figure of the distributed memory system nosql search caching that Fig. 5 provides for example IV.
Specific embodiment
For make present invention solves the technical problem that, the technical scheme that adopts and the technique effect that reaches clearer, below
By combine accompanying drawing the technical scheme of the embodiment of the present invention is described in further detail it is clear that described embodiment only
It is a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those skilled in the art exist
The every other embodiment being obtained under the premise of not making creative work, broadly falls into the scope of protection of the invention.
Further illustrate technical scheme below in conjunction with the accompanying drawings and by specific embodiment.Need explanation
It is that the present invention taking following examples as a example illustrates to technical scheme, but not in this, as restriction.This area
Technical staff can understand, optimization method and system that money distributed memory system nosql search proposed by the invention caches
In addition to for distributed memory system, can also be widely used in other same or like fields, and obtain similar skill
Art effect.
Under the conditions of data center environment, the Database Systems of bottom employed technical scheme comprise that hbase is distributed and deposit
Storage system, hbase is a storage system organizing bottom data by the way of lsm, and under the conditions of data center environment,
The application such as remote sensing satellite assumes data access locality and how inferior feature is once read in write.
Hereinafter the hbase storage system relevant technical terms used in embodiment are illustrated:
(1) hbase assembly: include hmaster, hregionserver and client in hbase storage system.Wherein
Hmaster is responsible for the operation such as the establishment of tables of data, deletion in hbase storage system.Concrete table is stored on hregionserver
Data, including multiple hregion of table.
(2) hregion: each table is divided into multiple hregion, each hregion comprises multiple store, each store pair
Ying Yuyi row cluster (column family), each store comprises a memstore and multiple storefile.
(3) compact and split: in hbase system operation, multiple hfile files can be produced, when hfile literary composition
When part number exceedes some, hbase cluster can execute compact.When hfile file size exceedes certain threshold values
Shi Zhihang split operates.
When one important characteristic of hbase storage system is in hfile file write hdfs cluster, hfile literary composition
Part no longer changes.
Embodiment one:
Fig. 1 is for being the optimization method flow chart of the distributed memory system nosql search caching that the present embodiment provides.Reference
Shown in Fig. 1, the method comprises the steps:
Step s101, in hbase client, pre-sets local ssd as only read buffer;
After in hbase client (distributed memory system cluster) as shown in Figure 2, the local ssd of setting is as only read buffer
Configuration diagram.Wherein local ssd can add in hregionserver.
Step s102, when hbase client reads target data, judges whether target data is located at described local ssd
On, if it is, entering step s104;If it is not, then entering step s103;
Before this step, hbase client produces get/scan operation requests according to demand, according to the number of targets of operation
According to, hbase client, the operation requests such as get/scan are sent to corresponding hregionserve, hregionserver according to
The read requests receiving, judge whether target data is located locally in internal memory, wherein assume to need to read hfile file, fixed
The specific hfile file in position, judges data base in described hfile file whether in local memory, if it is, from local
Described target data is returned, if not, i.e. corresponding hfileblock data block not in local memory, then enters step in internal memory
Rapid s102.
Step s103, reads target data from hdfs cluster and is cached on described local ssd;
Wherein, can also include before step s103: judge whether the residual memory space of described local ssd meets pre-
If requiring, if it is, entering step s103;If it is not, then execution replacement algorithm such as lru (least recently used,
Least recently used) algorithm, replace the data with existing in described local ssd with the target data reading, as with reading
Hfile file replaces the old hfile file in local ssd.
Step s104, described local ssd returns described target data.
Corresponding, described step s104 specifically may include that described local ssd return described target data arrive described locally
Internal memory, and described local memory returns described target data.
Additionally, in above-described embodiment, underlying file systems can be directly write to when writing data, at described
Row cache is not entered in ground ssd.
Under extensive mass data environment, by nosql storage system data storage, quick lookup is proposed low
Delay requirement.And the mode of the nosql system commonly used lsm tree with bigtable as representative is organizing semi-structured number at present
According to, for write once read multiple feature.
And the technical scheme that the present embodiment provides, introducing local ssd provides read-only caching function, give full play to ssd with
Machine reading performance, caches associated documents in ssd, and the data because accessing has locality, permissible by the file that caches in ssd
Effectively reducing the i/o number reading hdfs, thus improving the performance of distributed memory system, reaching raising data search efficiency
Etc. purpose.
In the application, for hbase, by the way of lsm tree, group organization data and upper strata reading data are many for technique scheme
Load, using local ssd as the read-only caching function of hbase cluster, make full use of the feature of hbase itself, especially suitable
In the application such as upper strata remote sensing satellite, data access is read to write few load characteristic more.
Embodiment two:
In hbase storage system running, multiple hfile files can be produced, when hfile file number exceedes necessarily
When number, hbase cluster needs to execute compact operation, to reduce hfile number of files.The present embodiment is in embodiment
On the basis of one optimization method of distributed memory system nosql search caching, there is provided one kind is in the execution of hbase cluster
Compact operational approach, as described in Figure 3 for the schematic flow sheet of the method, specifically includes following steps:
Step s301, hbase client generates compact operation requests;
Step s302, judges whether compact operation requests corresponding target hfile file is located on described local ssd,
If it is, entering step s304;If it is not, then entering step s303;
Step s303, reads target hfile file cache from hdfs cluster on described local ssd;
Step s304, execution compact operation;
Step s305, after the completion of compact operation, former hfile file is deleted from local ssd;
Step s306, newly-generated hfile file is write hdfs cluster and is cached in local ssd.
The method being provided by the present embodiment, can be executed when exceeding some when hfile file number
Compact operates, to reduce hfile number of files.
Embodiment three:
In hbase storage system running, multiple hfile files can be produced, when hfile file size exceedes necessarily
Threshold values when, hbase cluster need execute split operate to reduce hfile file size.The present embodiment embodiment one point
On the basis of the optimization method of cloth storage system nosql search caching, there is provided one kind is in hbase cluster execution split behaviour
Make method, as described in Figure 4 for the schematic flow sheet of the method, specifically include following steps:
Step s401, hbase client generates split operation requests;
Step s402, judges whether split operation requests corresponding target hfile file is located on described local ssd, such as
Fruit is then to enter step s304;If it is not, then entering step s303;
Step s403, reads target hfile file cache from hdfs cluster on described local ssd;
Step s404, execution split operation;
Step s405, after the completion of split operation, former hfile file is deleted from local ssd.
The method being provided by the present embodiment, can be executed when hfile file size exceedes certain threshold values
Split operates, to reduce hfile file size.
Example IV:
Corresponding to said method, the present embodiment additionally provides a kind of optimization of distributed memory system nosql search caching
System, this system architecture schematic diagram as shown in Figure 5, comprising:
Local ssd501, for as read-only buffer setting in hbase client;
First judge module 502, for when hbase client reads target data, judging whether target data is located at
On described local ssd;
Data read module 503, for when target data is not located on described local ssd, reading from hdfs cluster
Target data is simultaneously cached on described local ssd;
Data returns module 504, for when target data is located on described local ssd, returning institute from described local ssd
State target data.
Additionally, described system, can also include:
Second judge module, whether the residual memory space for judging described local ssd meets preset requirement;
Replacement module, when being unsatisfactory for preset requirement for the residual memory space in described local ssd, execution is replaced and is calculated
Method, such as lru (least recently used, least recently used), are replaced in described local ssd with the target data reading
Data with existing, such as replace the old hfile file in local ssd with the hfile file reading.
In hbase storage system running, multiple hfile files can be produced, when hfile file number exceedes necessarily
When number, hbase cluster needs to execute compact operation, to reduce hfile number of files.Therefore described system, also may be used
To include:
Compact module, for when target hfile file is located on described local ssd, execution compact operates, will
Former hfile file is deleted from local ssd, and by newly-generated hfile file write hdfs cluster and is cached to local ssd
In.
When hfile file size exceedes certain threshold values, hbase cluster needs to execute split operation to reduce hfile
File size.Therefore described system, can also include:
Split module, for when target hfile file is located on described local ssd, execution split operates, will be former
Hfile file is deleted from local ssd.
The technical scheme that the present embodiment provides, introducing local ssd provides read-only caching function, gives full play to the random of ssd
Reading performance, caches associated documents in ssd, and the data because accessing is had locality, can be had by the file caching in ssd
The i/o number of hdfs is read in the minimizing of effect, thus improving the performance of distributed memory system, reaching and improving data search efficiency
Etc. purpose.
In the application, for hbase, by the way of lsm tree, group organization data and upper strata reading data are many for technique scheme
Load, using local ssd as the read-only caching function of hbase cluster, make full use of the feature of hbase itself, especially suitable
In the application such as upper strata remote sensing satellite, data access is read to write few load characteristic more.
On software, operating system is preferably Linux system to the present invention, operates in and provides file io clothes in a linux group of planes
On the software of business, the such as nosql distributed data base system such as hdfs, gfs distributed file system and hbase, and
Hdfs distributed file system configures multiple datanode.
All or part of content in the technical scheme that above example provides can pass through software programming or specialized hardware
Equipment realize, wherein software program is stored in the storage medium that can read, storage medium for example: the hard disk in computer, light
Disk or floppy disk;Special hardware can be asic, fpga, soc or the ip core with related circuit.
Note, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore although being carried out to the present invention by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other Equivalent embodiments more can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of optimization method of distributed memory system nosql search caching is it is characterised in that include:
Step s101, in hbase client, pre-sets local ssd as only read buffer;
Step s102, when hbase client reads target data, judges whether target data is located on described local ssd, such as
Fruit is then to enter step s104;If it is not, then entering step s103;
Step s103, reads target data from hdfs cluster and is cached on described local ssd;
Step s104, described local ssd returns described target data.
2. the method for claim 1 is it is characterised in that before described step s102, also include:
Hbase client sends read requests to corresponding hregionserve;
Hregionserver, according to the read requests receiving, judges whether target data is located locally in internal memory, if it is,
Then return described target data from local memory, if it is not, then entering step s102.
3. method as claimed in claim 2 is it is characterised in that described step s104 includes:
Described local ssd returns described target data to described local memory, and described local memory returns described number of targets
According to.
4. the method for claim 1 is it is characterised in that before described step s103, also include:
Judge whether the residual memory space of described local ssd meets preset requirement, if it is, entering step s103;If
No, then algorithm is replaced in execution, replaces the data with existing in described local ssd with the target data reading.
5. the method for claim 1 is it is characterised in that also include:
Hbase client generates compact operation requests, whether judges compact operation requests corresponding target hfile file
On described local ssd, if it is, execution compact operation;If it is not, then reading target hfile from hdfs cluster
File cache and executes compact operation, after the completion of compact operation, by former hfile file from this on described local ssd
Delete in ground ssd, and by newly-generated hfile file write hdfs cluster and be cached in local ssd.
6. the method for claim 1 is it is characterised in that also include:
Hbase client generates split operation requests, judges whether split operation requests corresponding target hfile file is located at
On described local ssd, if it is, execution split operation;If it is not, then read target hfile file from hdfs cluster delaying
It is stored on described local ssd and executes split operation, after the completion of split operation, former hfile file is deleted from local ssd
Remove.
7. a kind of optimization system of distributed memory system nosql search caching is it is characterised in that include:
Local ssd, for as read-only buffer setting in hbase client;
First judge module, described local for when hbase client reads target data, judging whether target data is located at
On ssd;
Data read module, for when target data is not located on described local ssd, reading target data from hdfs cluster
And be cached on described local ssd;
Data returns module, for when target data is located on described local ssd, returning described target from described local ssd
Data.
8. system as claimed in claim 7 is it is characterised in that also include:
Second judge module, whether the residual memory space for judging described local ssd meets preset requirement;
Replacement module, when being unsatisfactory for preset requirement for the residual memory space in described local ssd, algorithm is replaced in execution, uses
The target data reading replaces the data with existing in described local ssd.
9. system as claimed in claim 7 is it is characterised in that also include:
Compact module, for when target hfile file is located on described local ssd, execution compact operates, will be former
Hfile file is deleted from local ssd, and by newly-generated hfile file write hdfs cluster and is cached to local ssd
In.
10. system as claimed in claim 7 is it is characterised in that also include:
Split module, for when target hfile file is located on described local ssd, execution split operation, by former hfile
File is deleted from local ssd.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610744493.9A CN106354805A (en) | 2016-08-28 | 2016-08-28 | Optimization method and system for searching and caching distribution storage system NoSQL |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610744493.9A CN106354805A (en) | 2016-08-28 | 2016-08-28 | Optimization method and system for searching and caching distribution storage system NoSQL |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106354805A true CN106354805A (en) | 2017-01-25 |
Family
ID=57856191
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610744493.9A Pending CN106354805A (en) | 2016-08-28 | 2016-08-28 | Optimization method and system for searching and caching distribution storage system NoSQL |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106354805A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107122264A (en) * | 2017-05-15 | 2017-09-01 | 成都优孚达信息技术有限公司 | mass data disaster-tolerant backup method |
CN107562385A (en) * | 2017-09-13 | 2018-01-09 | 郑州云海信息技术有限公司 | Distributed storage client reads the method, apparatus and equipment of data |
CN107632784A (en) * | 2017-09-14 | 2018-01-26 | 郑州云海信息技术有限公司 | The caching method of a kind of storage medium and distributed memory system, device and equipment |
CN107797771A (en) * | 2017-11-16 | 2018-03-13 | 郑州云海信息技术有限公司 | A kind of multipath storage optimization method and device |
CN109491789A (en) * | 2018-11-02 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | A kind of distributed memory system traffic balancing processing method, device and equipment |
CN109783032A (en) * | 2019-01-24 | 2019-05-21 | 山东超越数控电子股份有限公司 | A kind of distributed storage accelerating method and device based on Heterogeneous Computing |
CN111274310A (en) * | 2018-12-05 | 2020-06-12 | 中国移动通信集团山东有限公司 | Distributed data caching method and system |
CN111291040A (en) * | 2018-12-10 | 2020-06-16 | 中国移动通信集团四川有限公司 | Data processing method, device, equipment and medium |
CN112764690A (en) * | 2021-02-03 | 2021-05-07 | 北京同有飞骥科技股份有限公司 | Distributed storage system |
CN113835614A (en) * | 2020-09-17 | 2021-12-24 | 北京焱融科技有限公司 | SSD intelligent caching method and system based on distributed file storage client |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110173395A1 (en) * | 2010-01-12 | 2011-07-14 | International Business Machines Corporation | Temperature-aware buffered caching for solid state storage |
CN104267912A (en) * | 2014-09-19 | 2015-01-07 | 北京联创信安科技有限公司 | NAS (Network Attached Storage) accelerating method and system |
CN105446665A (en) * | 2015-12-18 | 2016-03-30 | 长城信息产业股份有限公司 | Computer storage acceleration system and optimization method thereof |
-
2016
- 2016-08-28 CN CN201610744493.9A patent/CN106354805A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110173395A1 (en) * | 2010-01-12 | 2011-07-14 | International Business Machines Corporation | Temperature-aware buffered caching for solid state storage |
CN104267912A (en) * | 2014-09-19 | 2015-01-07 | 北京联创信安科技有限公司 | NAS (Network Attached Storage) accelerating method and system |
CN105446665A (en) * | 2015-12-18 | 2016-03-30 | 长城信息产业股份有限公司 | Computer storage acceleration system and optimization method thereof |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107122264A (en) * | 2017-05-15 | 2017-09-01 | 成都优孚达信息技术有限公司 | mass data disaster-tolerant backup method |
CN107122264B (en) * | 2017-05-15 | 2020-06-09 | 成都优孚达信息技术有限公司 | Disaster-tolerant backup method for mass data |
CN107562385A (en) * | 2017-09-13 | 2018-01-09 | 郑州云海信息技术有限公司 | Distributed storage client reads the method, apparatus and equipment of data |
CN107632784A (en) * | 2017-09-14 | 2018-01-26 | 郑州云海信息技术有限公司 | The caching method of a kind of storage medium and distributed memory system, device and equipment |
CN107797771A (en) * | 2017-11-16 | 2018-03-13 | 郑州云海信息技术有限公司 | A kind of multipath storage optimization method and device |
CN109491789A (en) * | 2018-11-02 | 2019-03-19 | 浪潮电子信息产业股份有限公司 | A kind of distributed memory system traffic balancing processing method, device and equipment |
CN111274310A (en) * | 2018-12-05 | 2020-06-12 | 中国移动通信集团山东有限公司 | Distributed data caching method and system |
CN111291040A (en) * | 2018-12-10 | 2020-06-16 | 中国移动通信集团四川有限公司 | Data processing method, device, equipment and medium |
CN111291040B (en) * | 2018-12-10 | 2022-10-18 | 中国移动通信集团四川有限公司 | Data processing method, device, equipment and medium |
CN109783032A (en) * | 2019-01-24 | 2019-05-21 | 山东超越数控电子股份有限公司 | A kind of distributed storage accelerating method and device based on Heterogeneous Computing |
CN113835614A (en) * | 2020-09-17 | 2021-12-24 | 北京焱融科技有限公司 | SSD intelligent caching method and system based on distributed file storage client |
CN112764690A (en) * | 2021-02-03 | 2021-05-07 | 北京同有飞骥科技股份有限公司 | Distributed storage system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106354805A (en) | Optimization method and system for searching and caching distribution storage system NoSQL | |
US9672235B2 (en) | Method and system for dynamically partitioning very large database indices on write-once tables | |
CN110825748B (en) | High-performance and easily-expandable key value storage method by utilizing differentiated indexing mechanism | |
Liao et al. | Multi-dimensional index on hadoop distributed file system | |
US9286336B2 (en) | Unified architecture for hybrid database storage using fragments | |
CN102222085B (en) | Data de-duplication method based on combination of similarity and locality | |
CN105528367B (en) | Storage and near real-time querying method based on open source big data to time sensitive data | |
CN105912687B (en) | Magnanimity distributed data base storage unit | |
CN103812939A (en) | Big data storage system | |
CN104408111A (en) | Method and device for deleting duplicate data | |
CN110188080A (en) | Telefile Research of data access performance optimization based on client high-efficiency caching | |
US9348833B2 (en) | Consolidation for updated/deleted records in old fragments | |
CN103246616A (en) | Global shared cache replacement method for realizing long-short cycle access frequency | |
CN109521959A (en) | One kind being based on SSD-SMR disk mixing key assignments memory system data method for organizing | |
CN110058822A (en) | A kind of disk array transverse direction expanding method | |
CN105630919A (en) | Storage method and system | |
CN110309233A (en) | Method, apparatus, server and the storage medium of data storage | |
Lv et al. | Log-compact R-tree: an efficient spatial index for SSD | |
CN110109927A (en) | Oracle database data processing method based on LSM tree | |
CN102355502B (en) | Remote access method for remotely accessing storage system into desktop operation system | |
CN104572505A (en) | System and method for ensuring eventual consistency of mass data caches | |
CN104657461A (en) | File system metadata search caching method based on internal memory and SSD (Solid State Disk) collaboration | |
CN103942301A (en) | Distributed file system oriented to access and application of multiple data types | |
CN104158863A (en) | Cloud storage mechanism based on transaction-level whole-course high-speed buffer | |
CN116541427B (en) | Data query method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170125 |
|
RJ01 | Rejection of invention patent application after publication |